Study of deep learning models on educational channel video from YouTube for classification of Hinglish text

No Thumbnail Available
Sankar, Akshaya
Issue Date
MSc in Data Analytics
Dublin Business School
Items in eSource are protected by copyright. Previously published items are made available in accordance with the copyright policy of the publisher/copyright holder.
The applications of data mining process are widespread. Within the framework of CRISP-DM a data mining project, this research exposes key areas like data preparation, feature engineering, model training and evaluation techniques. Sentiment analysis / opinion mining is the technique for categorizing and defining computationally the opinions or feelings expressed in a piece of text to decide whether the attitude of individuals towards any topic, interest or product is a polarity of positive, negative or neutral. In numerous natural language processing assignments, deep learning is one of the most widely recognized methodologies. Deep learning models on normal language processing undertakings likewise beat regular AI models. It is a common practice these days for public to use social media to share their opinions and ideas about most of the things and topics related to education is one of the most common searches among many social media posts or videos. Everyday lot of educational videos are uploaded to YouTube platform and most people share their opinions about the channel or video in the comment section. The research focuses on how well deep learning techniques work in extracting student/viewer tendency of YouTubers from the Hinglish dataset. This research shall serve as a basis to provide useful information to teachers and online tutors by understanding and acting on the emotions of the students during the process of learning from their uploaded video and help them improve more on their methods of teaching.